Neuromelanin as a nidus for neurodegeneration

Abstract

This scientific commentary refers to ‘Neuromelanin accumulation drives endogenous synucleinopathy in non-human primates’ by Chocarro et al. (https://doi.org/10.1093/brain/awad331).

This scientific commentary refers to ‘Neuromelanin accumulation drives endogenous synucleinopathy in non-human primates’ by Chocarro et al. (https://doi.org/10.1093/brain/awad331).

The vulnerability of pigmented dopaminergic neurons in the substantia nigra pars compacta (SNc) has been known for many decades.¹ A comparison of neuronal loss in brainstem regions of individuals with idiopathic Parkinson’s disease and controls showed that the brain areas with the most neuromelanin-pigmented neurons were also those with the most degeneration.² Most neuromelanin-positive neurons are either dopaminergic or noradrenergic. Many brain areas positive for catecholamine neurotransmitters and neuromelanin can also harbour Lewy bodies, including the SNc, pedunculopontine nucleus, locus coeruleus, sympathetic ganglia, dorsal motor nucleus of the vagus, and the olfactory bulb. While pathological studies suggest a relationship between neuromelanin, Lewy pathology and neuronal death in Parkinson’s disease, because neuromelanin is not expressed in the brains of rodents or most primates, it has historically been difficult to prove that neuromelanin plays a role in neurodegenerative processes. In this issue of Brain, Chocarro and co-workers³ test the hypothesis that neuromelanin contributes to pathology in Parkinson’s disease by transducing the SNc of non-human primates with an adeno-associated viral vector (AAV) encoding human tyrosinase, the enzyme that catalyses formation of non-neuronal melanin in the periphery.

The main pathway for neuromelanin synthesis in the brain is via the conversion of oxidized dopamine (or norepinephrine) into quinones (Fig. 1).^4,5 First, tyrosine is converted to dihydroxyphenylalanine—otherwise known as L-DOPA—by tyrosine hydroxylase. L-DOPA is then converted to dopamine by the enzyme aromatic L-amino acid decarboxylase. Dopamine can, in turn, be oxidized by iron to form a dopamine (DA)-quinone, an aromatic ketone group, which is converted via subsequent pathways into pheomelanin and eumelanin, which polymerize to form neuromelanin. In the periphery, melanin is made via an enzymatic process in which tyrosine is converted to DOPA and DOPA-quinone by tyrosinase. Although this is not the main pathway for neuromelanin synthesis in neurons, low levels of tyrosinase are expressed in the brain and the contribution of enzymatic production of neuromelanin deserves further research.⁶ Once formed, neuromelanin is packed into autophagosomes and lysosomes where it likely performs a protective function by sequestering oxidized catecholamines, heavy metals, pesticides and other toxicants, as well as pathological α-synuclein aggregates. Mass spectrometry of neuromelanin-containing organelles from human SNc revealed the presence of proteins associated with autophagosomes, lysosomes and mitochondria, as well as lipid bodies composed of dolichols, glycerosphingolipids and phospholipids.⁷

Figure 1.

Neuromelanin and Parkinson’s disease. Expression of tyrosinase in dopaminergic neurons induces the formation of DOPA-quinones, which polymerize to form neuromelanin and recruit proteins including fibrillar α-synuclein, iron, toxicants and lipids. This complex is sequestered by autophagosomes, thereby helping to protect neurons from damage. In disease, neuromelanin may escape from lysosomes and cause neuronal damage or it may spread to neighbouring cells such as pyramidal neurons in the cortex as demonstrated by Chocarro et al.³ Created with BioRender.com.

Currently, many animal models of Parkinson’s disease are based on overexpression of α-synuclein, for example via AAV vectors, or on the use of preformed fibrils (PFFs), which induce α-synuclein to form aggregates. However, overexpression of α-synuclein does not replicate the expression levels seen in idiopathic Parkinson’s disease, and does not lead to the formation of inclusions similar to those of Parkinson’s disease; instead most of the α-synuclein is of normal conformation and soluble. The PFF model results in inclusions with many biochemical and morphological features of early α-synuclein aggregates, but only at later time points can Lewy body-like aggregates be found in a small percentage of neurons.⁸

Non-human primates provide important models of Parkinson’s disease because the basal ganglia circuitry and motor behaviours of primates more closely resemble those of humans. In their new model, Chocarro and colleagues³ therefore used an AAV vector to express human tyrosinase in the SNc of macaques. Tyrosinase expression induced pigmentation in the SNc that could be visualized at a gross anatomic level. Remarkably, tyrosinase expression led to the formation of Lewy-like pathology that was morphologically more similar to that of Parkinson’s disease than the pathology found in any other α-synuclein model to date. These findings are reminiscent of those in a previous study in which AAV-tyrosinase was injected into the SNc of rats.⁹ In rodents, exogenous expression of tyrosinase led to production of Lewy-like pathology and progressive loss of dopaminergic neurons, with up to 90% lost after 24 months. In macaques, tyrosinase expression led to a significant, ∼50% loss of dopaminergic terminals in the putamen, and loss of dopaminergic neurons in the SNc was observed at 4 and 8 months post-transduction, but only in female macaques. Interestingly, neuromelanin aggregates were found in pyramidal neurons of the cortex juxtaposed to dopaminergic terminals, suggesting that neuromelanin/α-synuclein complexes may be released and then taken up by neighbouring cells that do not express neuromelanin.

The use of macaques imposed constraints on the sample size in this study. In addition, the technical challenges associated with performing stereotaxic surgeries and injections into the SNc may have reduced the transduction efficiency of AAV-tyrosinase. However, the selective loss of SNc dopaminergic neurons in female macaques compared to males could point to important sex differences in the impact of neuromelanin on Parkinson’s disease pathology. These findings should be explored in future studies in both rodents and non-human primates.

Lysosome dysfunction, mitochondrial damage and α-synuclein pathology have all been linked to Parkinson’s disease. Neuromelanin may serve as a platform bringing lysosomes, mitochondrial oxidative stress, damaged lipids and fibrillar α-synuclein together to induce neuronal death (Fig. 1). Several questions remain, such as how neuromelanin-expressing cells replicate Lewy-like pathology better than cells in which α-synuclein expression alone has been manipulated experimentally. Examples include the sequestration of iron and oxidized dopamine to produce post-translational modifications of α-synuclein that promote fibrillization. Another question is how do neuromelanin complexes induce toxicity? Possible mechanisms could include lysosomal escape and release of toxins into the cytosol, or release of neuromelanin from neurons and activation of damaging inflammatory pathways. In addition, Lewy pathology can also be found in non-neuromelanin expressing neurons. What are the mechanisms by which α-synuclein becomes pathological in non-catecholaminergic neurons? Can neuromelanin inherited from neighbouring noradrenergic or dopaminergic processes template Lewy pathology in cortical neurons, for example? Exploring the mechanisms by which neuromelanin induces toxicity will help us discover ways of preventing Parkinson’s disease phenotypes.

Funding

This work was supported by National Insitute of Health, National Institute of Neurological Disorders and Stroke 1 R01 AG081433-01.

Competing interests

The author reports no competing interests.